The present study assesses the contribution of genetic and environmental factors to variability in placental aryl hydrocarbon hydroxylase and glutathione transferase activities using twin study methodology. Twin placentas were collected at the time of delivery. The placenta, except for a single layer of maternal decidua, consists of fetal tissue exhibiting fetal genotype. Microsomal and cytosolic fractions were prepared under stringent protocols to prevent enzyme activity loss. There were two monozygotic-monochorionic pairs, five monozygotic-dichorionic pairs, and 21 dizygotic-dichorionic pairs that showed measurable aryl hydrocarbon hydroxylase activity using the direct fluorometric assay. Most of the mothers were smokers. Aryl hydrocarbon hydroxylase activity was measured with two different substrates, benzo(a)pyrene and 7-ethoxyresorufin. Glutathione transferase activity was measured using glutathione and 1-chloro-2,4-dinitrobenzene as substrates for a spectrophotometric assay that follows the conversion of the aromatic substrate. Twin pair similarity was calculated with intraclass correlation coefficients. There is a high correlation between the activities of the two aryl hydrocarbon hydroxylase substrates (r = .814), but no correlation between aryl hydrocarbon hydroxylase and glutathione transferase activity levels. There is little evidence of genetic variability underlying the variation in the enzyme activities because monozygotic-dichorionic twins are no more similar to each other for the three substrate activities than are the dizygotic twins. To delineate the prenatal environmental influences on placental enzyme variability, dichorionic placentation was subdivided further into contiguous and noncontiguous placental position. Lower intraclass correlation coefficients are obtained for the dizygotic twins whose placentas were noncontiguous compared with dizygotic twins with contiguous placentas. The results suggest that most of the variability seen in these placental enzyme systems is due to environmental differences within uteri, rather than genetic variability in the population. This does not negate the possibility that between-pair, or population, variability may have a genetic component, because even dizygotic twins share a large proportion of their genes. This study points out that a significantly variable environment exists within the human uterus.